Artificial intelligence and autonomous operation are the most widely anticipated technologies in the on-road transportation sector, and with self-driving cars carrying paying passengers in several cities around the world, the promise is finally becoming reality.
The prime movers in the global economy however are the machines that build infrastructure and drive agriculture, and in the off-highway space, similar advanced technologies are revolutionizing the way earth is moved and crops are grown.
However, the challenge is much more considerable in the off-highway space. Tunneling, excavating, grading, tilling, and seeding are using GPS and laser guided systems to achieve unprecedented levels of accuracy and precision. What will the future look like for the off-highway equipment industry? How will advancements in hydraulic systems help achieve this future?
Jim Anderton discusses the future of off-highway equipment with Sara Feuling, Senior Director, Construction at the Association of Equipment Manufacturers (AEM) and Ben Holter, Director of Automation Systems for Husco.
Episode Transcript:
To see any graphics, images, and/or videos to which the transcript may be referring, please watch the above video. The transcript has been edited for clarity.
Jim Anderton: Hello everyone. And welcome to Manufacturing the Future. The transportation industry is undergoing a revolution driven by technology, such as battery electric drive, autonomous driving, and enabling technologies such as AI in the internet of things. But not everything that moves operates on roads. Agriculture and construction are cornerstones of the economy and that sector has enjoyed the same kind of technological revolution that's change in the automotive and aviation industries. In many cases, the demands are actually higher productivity, efficiency, safety, they're all important, and now environmental factors are entering the picture. Now, how is industry responding to this technological challenge? Joining me are Sara Feuling, Senior Director Construction at the Association of Equipment Manufacturers and Ben Holter, Director of Automation Systems for Husco. Sara holds bachelor's and master's degrees in civil engineering from the University of Wisconsin, Milwaukee and Ben holds a bachelor's degree in mechanical engineering from the Milwaukee School of Engineering, and a master's science degree from Northwestern University in predictive analytics. Sara and Ben, welcome to the show and I'm going to dive straight in because there's so much to talk about.
Autonomous vehicles, self-driving, everyone's talking about it. It's the hottest technology in the automotive industry today, and multiple companies are working on it. The trucking industry they're addressing labor shortages by adding sort of higher levels of technology, try and take the person out self-driving it's been talked about with various levels of success. There's just a ton of tech that is suddenly being adopted from what used to be sort of the software realm and we're jamming into things that traditionally were thought of as mechanical systems down there. What about OTR equipment? Will that self-driving loader greater backhoe, is that going to change everything? Are you going to get rid of the operator? Basically is someone from a beach and Cancun going to be going to be digging that trench? How about it, Sara? I'll start with you.
Sara Feuling: Those are all great points, Jim. We're seeing a lot in the on-road space and development and technology, and like you said, how all these pieces come together. We've seen a little bit of it already in mining, right? Where those more controlled operations are happening and that autonomy space is a little more fluid. But from an off-road perspective, we're absolutely bridging that gap now between what is typically heavy equipment, right? And technology. Part of it is just where we've come as an industry, right? But it's also part of that labor shortage is also a piece of it where we are able to use this technology to take what was a good operator and turn that person into a great operator.
There are automated features, machine control, grade control, all of those things that are really helping our operators in the field do more and do it better. They're eliminating rework and reducing the need for additional grade checks. And that's all of those advancements that we've seen in GPS, GNSS technology all kind of coming together. And we'll see where that takes us in the future. We definitely have a bright future ahead of us as we work through these things.
Jim Anderton: Sara, you're an expert in civil engineering and there's something I think that non-civil engineers have pondered. I certainly have, which is what defines accuracy and precision in your world compared to a world, for example, of machine tools or even 3D printing? I think there's sometimes a sense out there that, well, I work to tolerances of thousands of an inch and you're digging a tunnel. So, close enough is good enough. Just how tough is it to do those tasks?
Sara Feuling: We'll use something as simple as a grading operation, right? Tolerance is typically less than a 10th of a foot. So, that's an inch and a quarter, which sounds pretty big when you're relating it to something that's talking millimeters, right? But that inch and a quarter really adds up. And again, we rely on our strong experienced operators to get to that level of accuracy. But now when we combine that with the GPS and GNSS technology in grade control, we're able to bring that down to hundreds of feet. So, we're talking quarter inches and that in itself really kind of drives those efficiencies. You talked about that productivity as well as it all ends up to a dollar at the end of the day, right? So, if we're going to take that inch and a half inch, inch and a quarter over a 12 foot lane, every mile, we're almost at 240 cubic yards of material. So, now you're going to make that up in your pavement, you're looking at $35,000.
Not to the additional placement time, as well as the additional staff and work that requires. So, that accuracy that you talked about really adds up when you're able to use something like a grade control that's going to get you within those several hundredths of a foot versus a 10th of a foot. And now you're talking dozens of yards versus hundreds of yards.
Jim Anderton: Sara, do you still measure efficiency the same way that you used to in this industry? I mean, there's a used to be a sense of, I know what my operating cost per hour are, basically I've got a sense of how long it's going to take basically to sort of grade for that interstate. And I get a sense that time and cost overruns were fairly common, they still are. Is it tighter today?
Sara Feuling: I think definitely a lot of our contractors are getting smart. They're using the technology to their advantage, right? And they're finding those nuances, those productivity gains, those efficiency gains using some of the other tools, the software tools that are out there and available in the industry to really balance out and hammer out what those bid prices are, what that time is going to take, what that cost is going to take and in their operations are finding ways to get it done better and faster, which adds up to a dollar amount at the end of the day.
Jim Anderton: Ben, Sara talked about the idea of this advanced technology actually making a better operator or sort of a, of a super operator, if you will using sort of these devices. But equipment operators, they have specific tasks to perform in maneuvering vehicles and they have to operate multiple controls, all of us who've tried the incredibly complex technique of say operating even a backhoe are concerned. But good operators have an intuitive feel for things like rock and soil conditions, traction and gradient, breakout force, weight, they can sort of feel these things intuitively. Will future automation systems control these multiple factors in the way that human operators do? Do you have to put sensor systems in and then intelligent systems that operate the same way that a really good operator does when he just has an intuitive sense of what the critical lean angle is of the machine or what the breakout force is? Is it easy? Is it hard?
Ben Holter: Great question. Talking about the operator, an operator, a person is an amazing machine. It's really difficult to understand what's all going in our brain and how it operates. So, trying to put all of that into a control system is even more difficult. Automotive is running the same problems. How does machine vision work? How do you tell if it's a person or if it's a rock? If you can hit it, if you can dig it, or if it's dangerous? All of those things happen in the operator's brain instantly. And we know if it's right or wrong. It's really difficult to program and really difficult to understand. So, even breaking down just how to get back to that grade that Sara talked about; she talked about a 10th of a foot accuracy, that 10th of a foot at the bucket tip. If we talk about a 20-ton style excavator, that's on the side of the road, that 10th of a foot accuracy at the bucket tip equates back down to a one millimeter accuracy on the cylinder position.
That one millimeter accuracy is talking about the hydraulic compliance, the cylinder compliance, the hose compliance, even the structure compliance. If any of those stretch more than one millimeter or compressed more than one millimeter, now you're off by that one 10th of a foot. So, we think about how does a person do that? That's amazing. A person can control something that's that compliant to one millimeter. It's all about muscle memory. It's all about time in the seat and understanding how that works. We have amazing machine learning algorithms in our head that are constantly working that we don't even think about. And so, bringing that back down to how does a control system make that work? One millimeter on a cylinder, we're talking one milliamp on a control system to control a hydraulic control valve. That one milliamp is less than the parasitic of your phone charger sitting on the side of your bed right now.
So we think about these numbers and it just continues to drop orders of magnitude more and more tight, harder, and harder to achieve, but an operator's doing that every day. And so the difficulty we have as hydraulic manufacturers, as OEMs, as construction companies is how do we make something that maybe takes 10 or 20 or 30 years for an operator to get good at? How do we make that good day one? Because nobody's going to go buy an automated machine that it works like a two year old's driving it. Everybody wants to buy and operated an automated machine that operates as good as their 40 year experienced veteran today, day one. Because they've got the labor shortage, they want remote operation, they want all of these things that we have today with experienced operators. They want it a control system.
And so with the convergence of technology, GPS getting better, sensors getting better, more rugged for our industry. Dust, light vibration so much harsher in the construction industry that convergence of technology making it possible is now putting the onus on us hydraulic manufacturers and other controls manufacturers to make this possible. We are now able to do it and we now just have to prove that it's possible. And so really diving down into all the control systems, all the transfer functions of how each system needs to work and how tightly controlled each system needs to be is really the beginning of us making this work.
And we're starting to see in these really complex machines like excavators, starting to see this become more and more of a thing. We talk about automation, not autonomy, because the first step towards autonomy is automation. We need to automate some bits, whether it's grade or whether it's e- fencing. You talked about how we make the operators better. How can we do that? Well, make it so that if they're tired or not thinking that they can't swing into a road or they can't dig into a pipe.
And so if we have that information in the site plan, and we can program that into the machine, now we can make that happen. We can stop the operation of the excavator digging a pipe and having to move away everybody because there's a gas leak. But that all starts with electro hydraulics. A lot of machines don't even have electro hydraulics and you can't even plug in a control system to make that happen. And so we're incrementally taking steps forward in our industry to make all of this possible. And it's really an exciting time in our industry because all of this is becoming very possible and starting to show in the end market.
Jim Anderton: Ben you mentioned really the brilliance of the human body and how amazing it is that we have those proprioceptors and that incredible feedback system that lets us operate with that remarkable precision and touchdown. If you are going to duplicate that or enhance that with the systems you're developing at this point, what is your feedback system? What are you doing on the sensor side there? Is this radar, LIDAR, is this force sensing? Is this something you can detect, for example, back at the hydraulic cylinder? Or is this something that you need to do perhaps at the bucket or the end effector end? How do you even start this process?
Ben Holter: It's where do you start? Really, where you have to start is understanding that the operator today could look and see that grade has met, where it was supposed to meet. The GPS may tell it, or a surveyor may put a stake in the ground and tell the operator where the bucket tip is versus where it's supposed to be. So the first thing we need is sensors and those sensors have made huge strides and improvement in ruggedness and in accuracy. So now we're able to access satellites, access IMUs, inertial measurement systems on these machines to every 10 milliseconds or even faster in some cases, know where that bucket tip is and know where it is with relation to where it's supposed to be. So the first place we have to start is that. And tho that technology is finally becoming mature enough where it is rugged and we can rely it day after day, 24 hours a day, and start using those systems. To get where we're understanding things like loads, or soil conditions, or even obstacles such as people walking around.
We're still looking for advancements in sensors and reliability. Let's say there's 10 functions on a machine, if we add 20 sensors, one for each extended retract of a cylinder right at left of a motor. Now we've got 20 extra potential failure modes that could take down a system. Those extra potential failure modes could take down a machine, which could be on a path that's really important and could stop production of something for who knows how long, depending on how long it takes to get that machine back up and running or replace that sensor. So the more sensors we add, the more concern the operator's going to be for uptime.
And so we're always weighing what's the need of the sensor versus, the value of the sensor, versus what happens if the sensor goes wrong. So from a performance and accuracy of productivity and efficiency standpoint, then on top of that, a functional safety standpoint, we're always trying to weigh how many sensors do we add? Versus what does it give us? Versus what could go wrong in that state? And it's different for every customer, it's different for every system. And we're still trying to figure out what's the best solution. And it's going to continue to change as technology continues to advance the robustness, the acceptance of sensors, and the ability and the cost of sensors to put all of those on.
Jim Anderton: There used to be, it's anecdotally, or given a thought that junior operators, less experienced operators, are hard on equipment. And we know you could shock load hydraulic systems and create all sorts of problems back in the day when you were basically using, say fixed displacement, fixed pressure pumps, and basically with a blow off valve, and you're using accumulator to try and damp some of those, those shock loads. And a lot of things can happen, unexpected things can happen. And so in some ways you could define a great operator is one of those not only efficient, but also was easy on the machinery. We're looking at a world now where we may be bringing a lot more inexperienced people into the industry because of labor shortages at this point and getting them up to speed faster. Is this control technology, this kind of sensor technology, does that work into that factor too? Can that paper over some of the inexperience of an operator maybe even help them get better? Do you think?
Ben Holter: Yeah, it's a really exciting question because we're at a point right now where as electro hydraulics, as control systems on these large construction machines become more accepted, there are today in the automotive industry. We're able to add more sensors and add more controls to, like you said, help out this younger operators. But what that also does, which is really exciting from the hydraulic valve manufacturing design standpoint, is now we can start simplifying our hydraulic systems. The last six decades, we've continued to improve performance, improve efficiency, and improve controllability and safety by adding widgets because we didn't have a software control system that we could modify or change controls. It was a physical system. And so the only way to make it better is to add complexity, add extra widgets, to do the things that our customers are looking for.
Now that we can put software into this mix, we can start making our hydraulics simpler hydraulics, which means less failure points. But then we can give the same operation, the same functionality, productivity, efficiency that we're getting without all of those extra bits, all of those extra failure points. And then furthermore, we can modify how the valve feels depending on the operator. So we can put in a inexperienced operator mode that's maybe softer, more filtered, maybe lowered speeds. However, the end user or however, the owner.
However, the end user, however, the owner of the machine wants the operator act because in the end, if the operator's ruining the machine for the owner, they have to pay for it. And so we're able to do things depending on operator, depending on region, depending on what the machine's doing with the touch of a button, which has never been possibly before, which is what's making this so exciting for us in and beyond just automation.
Jim Anderton: Sara, in civil engineering, particularly big projects, we're seeing a revolution in the way some of these projects are designed and built. Now we're looking at things like undersea tunnels, very, very deep construction, skyscrapers, tall construction that involve foundation work in areas that were previously be considered completely unsuitable. In fact, new materials, new hybrid materials, concretes, not far from where I'm, we're talking about this, a new office towers going up, and it's about 12 stories it's made of wood of laminated wood, which is something which I never thought I would see in my life at all. Are new generations of machines going to be necessary in lockstep with this new pool of labor coming in to even build some of these projects. I mean, back in the day, if you bought a [Caterpillar] D9, it was expensive and you kept it on the job for 30 years you wore it out and you replaced it when you had to do, you have to now go through a different equipment cycle now based on its capability, rather than its life expectancy.
Sara Feuling: I don't know that it's necessarily a different life cycle, but there are those advancements to those machines. Let's take that D nine that you mentioned, right? You could have bought it 30 years ago, but all of our OEMs and our, and our technology providers and our hydraulic manufacturers are now making a lot of aftermarket retrofits, right? That you can bring your machine along with this advance in technology and these new operations, like you mentioned, there's different alternative materials, different construction methods, kind of changing how things are done, but it's not necessarily going to change your machine unless you need it to or want it to. So that's, again, the beauty of what of all our OEMs and our component suppliers are doing is they're giving end users and these contractors, the ability to build the machine that they need to best do their job. Whatever it is, right?
There's a lot of options out there. These technologies, and a lot of these control points that Ben was referencing, that all plays into what those OEMs are putting out for their big machine. And then the contractor, the end user can absolutely customize that. So obviously stuff that's coming off new, off the line has these things integrated, right? There's a lot of integral technology, which is great and wonderful to see, especially when we know this younger generation coming into the industry, they're used to computers, they're used to touch screens. They know what widget means, right? If you go back to some of them are experienced operators, those that have been in the industry for a while, when you're you're now instead of a couple controls, you've got touch screen. This younger generation is going to be able to just naturally get in line with that. Where I actually went up to the local here in Wisconsin, the operators union training center, and got to operate machine control grade control system. And you're still on the controls and the levers, but I got to sit in the seat and touch what was like an iPad and determine which mode I wanted it on, right? To make it easier for me, a very, very inexperienced operator to figure out how to use this technology, to do what it was that they wanted that wanted us to accomplish that day.
So I definitely do think there's, there's great opportunity for equipment, both new and existing to really align with where the future of construction is going.
Jim Anderton: So you mentioned specialty equipment. And that's a trend which we have noticed here at Engineering.com was in the past, if you were doing pipeline work for example, you used a basic track laying tractor chassis, and then you might have added a boom, for example, something simple as that. You might have added a grappler in place of a shovel, relatively simple sort of aftermarket sort of add-ons. Now we're seeing a whole generation of equipment now, which is incredibly specialized. Most recent. What I saw was actually a pavement milling machinery. Unusual stuff that does very, very special jobs, highly specialized in clearly not based as a modification of a simple existing chassis. Although clearly the underpinnings may come from a deer or a cat, or one of the OEMs at this point. Is that the future? Are we going to see a world in which there are smaller OEMs that are making very specialized equipment for a very small market just to get that last bit of productivity?
Sara Feuling: Yeah, absolutely. I think what we've seen a lot of actually is a lot of it is contractor driven. Where there are contractors who are very, very specialized and they rep they do on repetitive type of a specialized work and they have taken it on themselves to retrofit some sort of equipment or build their own attachment that does what two or three different attachments did previously. We are highlighting a lot of that. And you'll see, in CONEXPO CON/AGG 2023, we've got a lot of those features highlighted from an OEM perspective that you see these, these new solutions, right? They sound very simple. It's a common everyday problem, but it helps bring down the work, right, and the load and really specialized solutions to account for all those things.
Along with that, what Ben was mentioning and all those different sensors and all this hydraulic and where that's all moving, that's all going to be at IFPE as well. So within the industry itself, we are really highlighting some of those things, trying to draw attention to not only the industry, but to those smaller manufacturers, right? That, that really have a game changer, really have some stuff that's going to change the future of the industry and give them an opportunity to share that on a wider scale.
Ben Holter: And following up with Sara there, she talked about auxiliary and attachments. I think agriculture has done a great job. You have one power plant, you've got the tractor buy a ton of attachments and do what you need. We're starting to see that in our industry a lot more, especially at the supply side where ox, auxiliary attachments, auxiliary sections used to be just an add-on that you just needed to do a certain flow rate. Now, auxiliaries are becoming some of the most deluxe systems that we're selling the deluxe sections, because they have to have pressure control, flow control, a damping on them, zero back pressure for hammers. And, and we're putting a lot of the deluxe nest in those to handle all of the different auxiliary requirements that are being driven by end users that have this new attachment or this new attachment that never thought about before to get more, use more productivity, more uptime of a machine that's on a site rather than bringing in specialized machines for pipeline.
Now, like you said, you add an attachment to make it a bite layer. And that's really exciting too, because then you talk about software and what you can do with software. Now you talk about different pressures, different flow rates, rather than somebody having to go out and change a relief valve setting, or worry about blowing up their attachment, because it wasn't rated for that pressure. We can do that at, at, at a touch of a button to the hydraulic valve to change that setting and say, oh, I've now got a hammer or a grapple or a brush cutter. And we can set that all automatically to again, update productivity and also make it easier for new operators. So they don't have to worry about remembering to change relief valve setting before they blow up that brush cutter. So again, more opportunities for us to meet the market needs with hydraulic improvements that we've been waiting for to come.
Jim Anderton: Sara and Ben, do young engineering grads, students, perhaps the public at large. Are they aware of how much technology is in this industry? I think there's a conception out there that a bulldozer's a bulldozer it's been the same for half a century. I mean, we're, you're talking about cutting edge technology here is there a perception that it's there? Sara? How about you?
Sara Feuling: I definitely think it's not as out there as we would like it to be. Where I think you're exactly right, Jim, you look at a piece of machinery and you're like, "Ah, it's heavy machinery. It's dirty, it's basic, it's rudimentary," right? And it is so not. So we are, I think, as an industry, trying to help change that narrative, change that perception, tell that story. To say, "Look at all this technology and all these different things that are within our industries."
For example, if we're talking about all this software that goes into running these machines, somebody going to college right now, thinking they want to be a software engineer, isn't considering a career in construction, but they could be, and they should be. None of these things are possible without all of that technology and those typically non-construction or manufacturing roles. We are really broadening and widening that net of what is our future workforce pipeline. So as an association, as AEM, that is part of our workforce initiative, to help tell those stories, to tell some of these fun, exciting things that we're doing in this industry with technology to help draw the next generation and to change that perception.
Jim Anderton: Ben, your education could have led you to Silicon Valley.
Ben Holter: Yeah.
Jim Anderton: And how much fun are you having in this industry? This industry's clearly a place where high tech lives.
Ben Holter: It's so exciting. We just finished deploying and getting customer approval on a 10-function valve. And so you think about 10 functions, I've got one power plant that's controlling the power to 10 functions. And you think about the control system and how you control that. You've got 10 degrees of freedom off one power plant, all at different pressures, all at different forces that an operator expects to work harmoniously. And you don't think about how hard that is, because it's worked in the past hydromechanically, but when you put it into software and try and, again, make the hydromechanics a little bit more simple, it's a really complex situation.
It's more than just a gas pedal and a steering wheel, not to detriment anything that the other industries are doing, but it's an amazing problem. It's really exciting and really hard. It's all non-linear, it's all Bernoulli's Equations 10 times. And you think, well, hydraulics are going away, but then you think that that bucket cylinder, that's 20 feet out on a big excavator, that needs a hundred kilowatts to work. Imagine putting a hundred kilowatt electric motor diesel engine at the bucket cylinder to make that a distributed electronic system. You need a motor to get that power out and the power density of hydraulics are just so good and the robustness of hydraulics that it's going to be around for a while.
And so we're solving really exciting, hard problems in an industry that's being disrupted in an exciting way for us to solve global socioeconomic and geopolitical problems along the way. So it couldn't be more exciting and yeah, love what AM's doing to help make this initiative, to get people into this industry because hydraulics aren't the dirty, messy hydraulics you thought about 20 years ago. If you take a tour of our manufacturing plant, if you see us at FP, it's very clean, it's automated. Our valves are being manufactured in automated way, similar to what automotive is doing and what their industry's driven. So it's new, it's exciting, it's clean, and it's really hard problems that are, again, I'll just keep saying it: exciting.
Jim Anderton: There's just so much to talk about, we could do this for hours. I hope we get a chance to do more of this, but just to wrap this up or give a cap on this whole initial shallow dive that we've made in this topic down here, we're looking at a world now where the demographic challenges are real. This isn't a short-term blip where we just have to wait five years, we'll have another generation of skilled operators that will pop up out the woodwork. At the same time, the demands of the civil engineering, Sara, as you're noting, are much higher. We got to dig deeper, we got to do it cheaper, faster, new materials, all sorts of new factors at the same time. Look forward please. 10, 20, 30, 40 years from now basically, is this going to be the Jetsons? Are we going to have technology we don't even realize today? Or will this be something where this is recognizable but much different? Sara, I'll start with you.
Sara Feuling: I think that's a very interesting question, Jim, because you can look at it a couple of different ways, especially on the construction side. We're an industry typically very slow to change. We've done what we've done and we've done it well for years, but if you look at where we've come in the last 20 years and where we could potentially go, I like to always compare it to music.
Okay, so those of us that were born in the eighties, in my lifetime, from middle school, through high school, I went from a cassette to a CD, to an iPad, to my phone, in a 10 year window. Now extrapolate that and where the industry could be from here, right? We've got Tesla talking about being on Mars in 30 years, right? What are we going to be doing here?
So I think absolutely, there is the potential for 10 years from now, these advancements. And as that's proven, I think that's another thing that's going to - I think Ben mentioned it earlier, it needs to work and it needs to be a proven solution and a proven technology. And once the industry starts seeing that these things do work and they do the job well, I think at that point we'll absolutely see, 10, 20 years from now, these things will extrapolate. Tie that into these automated features that we've got, right? Everything we've talked about is automated, we're not quite too autonomy with any of these things, but this is step number one to get us to that potentially autonomous future. Pulling out some of those operators, pulling out some of the man hours on site, and being able to do this with the technology we have available.
So I think absolutely, it's very exciting to see where this could go and it could look very similar, but it also could look completely different. And that's for the industry to really help us drive towards that and realize what that shared future could be.
Jim Anderton: Ben, you'll be engineering that future. What do you think it'll look like?
Ben Holter: Yeah, putting machines and digging - putting them on Mars will be interesting - but we'll figure out a solution to do it. So that I think is what is cool about this industry, is that we need to be there to make sure that we're building houses, building roads, building infrastructure here, wherever it might be, speaking 40 years out. And we're going to continue to be more productive with it. And more safe with it. All the things we're doing are making people more safe, machines more safe, more productive, more efficient. And so the opportunities of where we can go in even 10 years are crazy to think about because there is just so much opportunity and with the convergence of all of the sensors and computing power and algorithms, really the opportunity is endless of what can our end users imagine as possible, and then we'll figure out a way to solve it. And that's what's exciting.
Jim Anderton: Sara Feuling, Association of Equipment Manufacturers; Ben Holter, Husco, thanks for joining me on the show.
Join us March 14-18 in Las Vegas at IFPE 2023. Use promo code DIG20 to save 20% off registration!